section 1 properties of structural steels …nguyen.hong.hai.free.fr/ebooks/science and...

1.1

SECTION 1PROPERTIES OF STRUCTURALSTEELS AND EFFECTS OFSTEELMAKING AND FABRICATION

R. L. Brockenbrough, P.E.President, R. L. Brockenbrough & Associates, Inc.,Pittsburgh, Pennsylvania

This section presents and discusses the properties of structural steels that are of importancein design and construction. Designers should be familiar with these properties so that theycan select the most economical combination of suitable steels for each application and usethe materials efficiently and safely.

In accordance with contemporary practice, the steels described in this section are giventhe names of the corresponding specifications of ASTM, 100 Barr Harbor Dr., West Con-shohocken, PA, 19428. For example, all steels covered by ASTM A588, ‘‘Specification forHigh-strength Low-alloy Structural Steel,’’ are called A588 steel.

1.1 STRUCTURAL STEEL SHAPES AND PLATES

Steels for structural uses may be classified by chemical composition, tensile properties, andmethod of manufacture as carbon steels, high-strength low-alloy steels (HSLA), heat-treatedcarbon steels, and heat-treated constructional alloy steels. A typical stress-strain curve for asteel in each classification is shown in Fig. 1.1 to illustrate the increasing strength levelsprovided by the four classifications of steel. The availability of this wide range of specifiedminimum strengths, as well as other material properties, enables the designer to select aneconomical material that will perform the required function for each application.

Some of the most widely used steels in each classification are listed in Table 1.1 withtheir specified strengths in shapes and plates. These steels are weldable, but the weldingmaterials and procedures for each steel must be in accordance with approved methods. Weld-ing information for each of the steels is available from most steel producers and inpublications of the American Welding Society.

1.1.1 Carbon Steels

A steel may be classified as a carbon steel if (1) the maximum content specified for alloyingelements does not exceed the following: manganese—1.65%, silicon—0.60%, copper—

1.2 SECTION ONE

FIGURE 1.1 Typical stress-strain curves for structural steels. (Curves havebeen modified to reflect minimum specified properties.)

0.60%; (2) the specified minimum for copper does not exceed 0.40%; and (3) no minimumcontent is specified for other elements added to obtain a desired alloying effect.

A36 steel is the principal carbon steel for bridges, buildings, and many other structuraluses. This steel provides a minimum yield point of 36 ksi in all structural shapes and inplates up to 8 in thick.

A573, the other carbon steel listed in Table 1.1, is available in three strength grades forplate applications in which improved notch toughness is important.

1.1.2 High-Strength Low-Alloy Steels

Those steels which have specified minimum yield points greater than 40 ksi and achieve thatstrength in the hot-rolled condition, rather than by heat treatment, are known as HSLA steels.Because these steels offer increased strength at moderate increases in price over carbon steels,they are economical for a variety of applications.

A242 steel is a weathering steel, used where resistance to atmospheric corrosion is ofprimary importance. Steels meeting this specification usually provide a resistance to atmos-pheric corrosion at least four times that of structural carbon steel. However, when required,steels can be selected to provide a resistance to atmospheric corrosion of five to eight timesthat of structural carbon steels. A specified minimum yield point of 50 ksi can be furnishedin plates up to 3⁄4 in thick and the lighter structural shapes. It is available with a lower yieldpoint in thicker sections, as indicated in Table 1.1.

A588 is the primary weathering steel for structural work. It provides a 50-ksi yield pointin plates up to 4 in thick and in all structural sections; it is available with a lower yield pointin thicker plates. Several grades are included in the specification to permit use of variouscompositions developed by steel producers to obtain the specified properties. This steel pro-vides about four times the resistance to atmospheric corrosion of structural carbon steels.

PROPERTIES OF STRUCTURAL STEELS AND EFFECTS OF STEELMAKING AND FABRICATION 1.3

TABLE 1.1 Specified Minimum Properties for Structural Steel Shapes and Plates*

ASTMdesignation

Plate-thicknessrange, in

ASTMgroup forstructuralshapes†

Yieldstress,ksi‡

Tensilestrength,

ksi‡

Elongation, %

In 2in§

In8 in

A36 8 maximum 1–5 36 58–80 23–21 20over 8 1–5 32 58–80 23 20

A573Grade 58 11⁄2 maximum � 32 58–71 24 21Grade 65 11⁄2 maximum � 35 65–77 23 20Grade 70 11⁄2 maximum � 42 70–90 21 18

High-strength low-alloy steels

A242 3⁄4 maximum 1 and 2 50 70 21 18Over 3⁄4 to 11⁄2 max 3 46 67 21 18Over 11⁄2 to 4 max 4 and 5 42 63 21 18

A588 4 maximum 1–5 50 70 21 18Over 4 to 5 max 1–5 46 67 21 —Over 5 to 8 max 1–5 42 63 21 —

A572Grade 42 6 maximum 1–5 42 60 24 20Grade 50 4 maximum 1–5 50 65 21 18Grade 60 11⁄4 maximum 1–3 60 75 18 16Grade 65 11⁄4 maximum 1–3 65 80 17 15

A992 � 1–5 50–65 65 21 18

Heat-treated carbon and HSLA steels

A633Grade A 4 maximum � 42 63–83 23 18Grade C, D 21⁄2 maximum � 50 70–90 23 18

Over 21⁄2 to 4 max � 46 65–85 23 18Grade E 4 maximum � 60 80–100 23 18

Over 4 to 6 max � 55 75–95 23 18

A678Grade A 11⁄2 maximum � 50 70–90 22 —Grade B 21⁄2 maximum � 60 80–100 22 —Grade C 3⁄4 maximum � 75 95–115 19 —

Over 3⁄4 to 11⁄2 max � 70 90–110 19 —Over 11⁄2 to 2 max � 65 85–105 19 —

Grade D 3 maximum � 75 90–110 18 —

A852 4 maximum � 70 90–110 19 —

A913 � 1–5 50 65 21 18� 1–5 60 75 18 16� 1–5 65 80 17 15� 1–5 70 90 16 14

1.4 SECTION ONE

TABLE 1.1 Specified Minimum Properties for Structural Steel Shapes and Plates* (Continued )

ASTMdesignation

Plate-thicknessrange, in

ASTMgroup forstructuralshapes†

Yieldstress,ksi‡

Tensilestrength,

ksi‡

Elongation, %

In 2in§

In8 in

Heat-treated constructional alloy steels

A514 21⁄2 maximum � 100 110–130 18 —Over 21⁄2 to 6 max � 90 100–130 16 —

* The following are approximate values for all the steels:Modulus of elasticity—29 � 103 ksi.Shear modulus—11 � 103 ksi.Poisson’s ratio—0.30.Yield stress in shear—0.57 times yield stress in tension.Ultimate strength in shear—2⁄3 to 3⁄4 times tensile strength.Coefficient of thermal expansion—6.5 � 10�6 in per in per deg F for temperature range �50 to �150�F.Density—490 lb / ft3.

† See ASTM A6 for structural shape group classification.‡ Where two values are shown for yield stress or tensile strength, the first is minimum and the second is maximum.§ The minimum elongation values are modified for some thicknesses in accordance with the specification for the

steel. Where two values are shown for the elongation in 2 in, the first is for plates and the second for shapes.� Not applicable.

These relative corrosion ratings are determined from the slopes of corrosion-time curvesand are based on carbon steels not containing copper. (The resistance of carbon steel toatmospheric corrosion can be doubled by specifying a minimum copper content of 0.20%.)Typical corrosion curves for several steels exposed to industrial atmosphere are shown inFig. 1.2.

For methods of estimating the atmospheric corrosion resistance of low-alloy steels basedon their chemical composition, see ASTM Guide G101. The A588 specification requires thatthe resistance index calculated according to Guide 101 shall be 6.0 or higher.

A588 and A242 steels are called weathering steels because, when subjected to alternatewetting and drying in most bold atmospheric exposures, they develop a tight oxide layerthat substantially inhibits further corrosion. They are often used bare (unpainted) where theoxide finish that develops is desired for aesthetic reasons or for economy in maintenance.Bridges and exposed building framing are typical examples of such applications. Designersshould investigate potential applications thoroughly, however, to determine whether a weath-ering steel will be suitable. Information on bare-steel applications is available from steelproducers.

A572 specifies columbium-vanadium HSLA steels in four grades with minimum yieldpoints of 42, 50, 60, and 65 ksi. Grade 42 in thicknesses up to 6 in and grade 50 inthicknesses up to 4 in are used for welded bridges. All grades may be used for riveted orbolted construction and for welded construction in most applications other than bridges.

A992 steel was introduced in 1998 as a new specification for rolled wide flange shapesfor building framing. It provides a minimum yield point of 50 ksi, a maximum yield pointof 65 ksi, and a maximum yield to tensile ratio of 0.85. These maximum limits are considereddesirable attributes for seismic design. To enhance weldability, a maximum carbon equivalentis also included, equal to 0.47% for shape groups 4 and 5 and 0.45% for other groups. Asupplemental requirement can be specified for an average Charpy V-notch toughness of 40ft � lb at 70�F.


FIGURE 1.2 Corrosion curves for structural steels in an industrial atmosphere. (From R. L.Brockenbrough and B. G. Johnston, USS Steel Design Manual, R. L. Brockenbrough & Associates,Inc., Pittsburgh, Pa., with permission.)

1.1.3 Heat-Treated Carbon and HSLA Steels

Both carbon and HSLA steels can be heat treated to provide yield points in the range of 50to 75 ksi. This provides an intermediate strength level between the as-rolled HSLA steelsand the heat-treated constructional alloy steels.

A633 is a normalized HSLA plate steel for applications where improved notch toughnessis desired. Available in four grades with different chemical compositions, the minimum yieldpoint ranges from 42 to 60 ksi depending on grade and thickness.

A678 includes quenched-and-tempered plate steels (both carbon and HSLA compositions)with excellent notch toughness. It is also available in four grades with different chemicalcompositions; the minimum yield point ranges from 50 to 75 ksi depending on grade andthickness.

A852 is a quenched-and-tempered HSLA plate steel of the weathering type. It is intendedfor welded bridges and buildings and similar applications where weight savings, durability,and good notch toughness are important. It provides a minimum yield point of 70 ksi inthickness up to 4 in. The resistance to atmospheric corrosion is typically four times that ofcarbon steel.

A913 is a high-strength low-allow steel for structural shapes, produced by the quenchingand self-tempering (QST) process. It is intended for the construction of buildings, bridges,and other structures. Four grades provide a minimum yield point of 50 to 70 ksi. Maximumcarbon equivalents to enhance weldability are included as follows: Grade 50, 0.38%; Grade60, 0.40%; Grade 65, 0.43%; and Grade 70, 0.45%. Also, the steel must provide an averageCharpy V-notch toughness of 40 ft � lb at 70�F.

1.1.4 Heat-Treated Constructional Alloy Steels

Steels that contain alloying elements in excess of the limits for carbon steel and are heattreated to obtain a combination of high strength and toughness are termed constructional

1.6 SECTION ONE

alloy steels. Having a yield strength of 100 ksi, these are the strongest steels in generalstructural use.

A514 includes several grades of quenched and tempered steels, to permit use of variouscompositions developed by producers to obtain the specified strengths. Maximum thicknessranges from 11⁄4 to 6 in depending on the grade. Minimum yield strength for plate thicknessesover 21⁄2 in is 90 ksi. Steels furnished to this specification can provide a resistance to at-mospheric corrosion up to four times that of structural carbon steel depending on the grade.

Constructional alloy steels are also frequently selected because of their ability to resistabrasion. For many types of abrasion, this resistance is related to hardness or tensile strength.Therefore, constructional alloy steels may have nearly twice the resistance to abrasion pro-vided by carbon steel. Also available are numerous grades that have been heat treated toincrease the hardness even more.

1.1.5 Bridge Steels

Steels for application in bridges are covered by A709, which includes steel in several of thecategories mentioned above. Under this specification, grades 36, 50, 70, and 100 are steelswith yield strengths of 36, 50, 70, and 100 ksi, respectively. (See also Table 11.28.)

The grade designation is followed by the letter W, indicating whether ordinary or highatmospheric corrosion resistance is required. An additional letter, T or F, indicates thatCharpy V-notch impact tests must be conducted on the steel. The T designation indicatesthat the material is to be used in a non-fracture-critical application as defined by AASHTO;the F indicates use in a fracture-critical application.

A trailing numeral, 1, 2, or 3, indicates the testing zone, which relates to the lowestambient temperature expected at the bridge site. (See Table 1.2.) As indicated by the firstfootnote in the table, the service temperature for each zone is considerably less than theCharpy V-notch impact-test temperature. This accounts for the fact that the dynamic loadingrate in the impact test is more severe than that to which the structure is subjected. Thetoughness requirements depend on fracture criticality, grade, thickness, and method of con-nection.

A709-HPS70W, designated as a High Performance Steel (HPS), is also now available forhighway bridge construction. This is a weathering plate steel, designated HPS because itpossesses superior weldability and toughness as compared to conventional steels of similarstrength. For example, for welded construction with plates over 21⁄2 in thick, A709-70Wmust have a minimum average Charpy V-notch toughness of 35 ft � lb at �10�F in Zone III,the most severe climate. Toughness values reported for some heats of A709-HPS70W havebeen much higher, in the range of 120 to 240 ft � lb at �10�F. Such extra toughness providesa very high resistance to brittle fracture.

(R. L. Brockenbrough, Sec. 9 in Standard Handbook for Civil Engineers, 4th ed., F. S.Merritt, ed., McGraw-Hill, Inc., New York.)

1.2 STEEL-QUALITY DESIGNATIONS

Steel plates, shapes, sheetpiling, and bars for structural uses—such as the load-carryingmembers in buildings, bridges, ships, and other structures—are usually ordered to the re-quirements of ASTM A6 and are referred to as structural-quality steels. (A6 does notindicate a specific steel.) This specification contains general requirements for delivery relatedto chemical analysis, permissible variations in dimensions and weight, permissible imper-fections, conditioning, marking and tension and bend tests of a large group of structuralsteels. (Specific requirements for the chemical composition and tensile properties of these


TABLE 1.2 Charpy V-Notch Toughness for A709 Bridge Steels*

Grade

Maximumthickness, in,

inclusive

Joining /fasteningmethod

Minimum averageenergy,

ft�lb

Test temperature, �F

Zone1

Zone2

Zone3

Non-fracture-critical members

36T 4 Mech. /Weld. 15 70 40 10

50T,†50WT†

2 Mech. /Weld. 15

2 to 4 Mechanical 15 70 40 102 to 4 Welded 20

70WT‡ 21⁄2 Mech. /Weld. 2021⁄2 to 4 Mechanical 20 50 20 �1021⁄2 to 4 Welded 25

100T,100WT

21⁄2 Mech. /Weld. 25

21⁄2 to 4 Mechanical 25 30 0 �3021⁄2 to 4 Welded 35

Fracture-critical members

36F 4 Mech. /Weld.a 25 70 40 10

50F,† 50WF† 2 Mech. /Weld.a 25 70 40 102 to 4 Mechanicala 25 70 40 102 to 4 Weldedb 30 70 40 10

70WF‡ 21⁄2 Mech. /Weld.b 30 50 20 �1021⁄2 to 4 Mechanicalb 30 50 20 �1021⁄2 to 4 Weldedc 35 50 20 �10

100F, 100WF 21⁄2 Mech. /Weld.c 35 30 0 �3021⁄2 to 4 Mechanicalc 35 30 0 �3021⁄2 to 4 Weldedd 45 30 0 NA

* Minimum service temperatures:Zone 1, 0�F; Zone 2, below 0 to �30�F; Zone 3, below �30 to �60�F.

† If yield strength exceeds 65 ksi, reduce test temperature by 15�F for each 10 ksi above 65 ksi.‡ If yield strength exceeds 85 ksi, reduce test temperature by 15�F for each 10 ksi above 85 ksi.a Minimum test value energy is 20 ft-lb.b Minimum test value energy is 24 ft-lb.c Minimum test value energy is 28 ft-lb.d Minimum test value energy is 36 ft-lb.

steels are included in the specifications discussed in Art. 1.1.) All the steels included in Table1.1 are structural-quality steels.

In addition to the usual die stamping or stenciling used for identification, plates and shapesof certain steels covered by A6 are marked in accordance with a color code, when specifiedby the purchaser, as indicated in Table 1.3.

Steel plates for pressure vessels are usually furnished to the general requirements ofASTM A20 and are referred to as pressure-vessel-quality steels. Generally, a greater numberof mechanical-property tests and additional processing are required for pressure-vessel-quality steel.

1.8 SECTION ONE

TABLE 1.3 Identification Colors

Steels Color Steels Color

A36 None A913 grade 50 red and yellow

A242 Blue A913 grade 60 red and gray

A514 Red A913 grade 65 red and blue

A572 grade 42 Green and white A913 grade 70 red and white

A572 grade 50 Green and yellow

A572 grade 60 Green and gray

A572 grade 65 Green and blue

A588 Blue and yellow

A852 Blue and orange

1.3 RELATIVE COST OF STRUCTURAL STEELS

Because of the many strength levels and grades now available, designers usually must in-vestigate several steels to determine the most economical one for each application. As aguide, relative material costs of several structural steels used as tension members, beams,and columns are discussed below. The comparisons are based on cost of steel to fabricators(steel producer’s price) because, in most applications, cost of a steel design is closely relatedto material costs. However, the total fabricated and erected cost of the structure should beconsidered in a final cost analysis. Thus the relationships shown should be considered asonly a general guide.

Tension Members. Assume that two tension members of different-strength steels have thesame length. Then, their material-cost ratio C2 /C1 is

C A p2 2 2� (1.1)C A p1 1 1

where A1 and A2 are the cross-sectional areas and p1 and p2 are the material prices per unitweight. If the members are designed to carry the same load at a stress that is a fixed per-centage of the yield point, the cross-sectional areas are inversely proportional to the yieldstresses. Therefore, their relative material cost can be expressed as

FC py12 2� (1.2)C F p1 y2 1

where Fy1 and Fy2 are the yield stresses of the two steels. The ratio p2 /p1 is the relative pricefactor. Values of this factor for several steels are given in Table 1.4, with A36 steel as thebase. The table indicates that the relative price factor is always less than the correspondingyield-stress ratio. Thus the relative cost of tension members calculated from Eq. (1.2) favorsthe use of high-strength steels.

Beams. The optimal section modulus for an elastically designed I-shaped beam resultswhen the area of both flanges equals half the total cross-sectional area of the member.Assume now two members made of steels having different yield points and designed to carrythe same bending moment, each beam being laterally braced and proportioned for optimal


TABLE 1.4 Relative Price Factors*

Steel

Minimumyield

stress, ksi

Relativepricefactor

Ratio ofminimum

yieldstresses

Relativecost oftension

members

A36 36 1.00 1.00 1.00

A572 grade 42 42 1.09 1.17 0.93

A572 grade 50 50 1.12 1.39 0.81

A588 grade A 50 1.23 1.39 0.88

A852 70 1.52 1.94 0.78

A514 grade B 100 2.07 2.78 0.75

* Based on plates 3⁄4 � 96 � 240 in. Price factors for shapes tend to be lower.A852 and A514 steels are not available in shapes.

section modulus. Their relative weight W2 /W1 and relative cost C2 /C1 are influenced by theweb depth-to-thickness ratio d / t. For example, if the two members have the same d / t values,such as a maximum value imposed by the manufacturing process for rolled beams, therelationships are

2 / 3FW y12 � (1.3)� �W F1 y2

2 / 3FC p y12 2� (1.4)� �C p F1 1 y2

If each of the two members has the maximum d / t value that precludes elastic web buckling,a condition of interest in designing fabricated plate girders, the relationships are

1 / 2FW y12 � (1.5)� �W F1 y2

1 / 2FC p y12 2� (1.6)� �C p F1 1 y2

Table 1.5 shows relative weights and relative material costs for several structural steels.These values were calculated from Eqs. (1.3) to (1.6) and the relative price factors given inTable 1.4, with A36 steel as the base. The table shows the decrease in relative weight withincrease in yield stress. The relative material costs show that when bending members arethus compared for girders, the cost of A572 grade 50 steel is lower than that of A36 steel,and the cost of other steels is higher. For rolled beams, all the HSLA steels have marginallylower relative costs, and A572 grade 50 has the lowest cost.

Because the comparison is valid only for members subjected to the same bending moment,it does not indicate the relative costs for girders over long spans where the weight of themember may be a significant part of the loading. Under such conditions, the relative materialcosts of the stronger steels decrease from those shown in the table because of the reductionin girder weights. Also, significant economies can sometimes be realized by the use of hybridgirders, that is, girders having a lower-yield-stress material for the web than for the flange.HSLA steels, such as A572 grade 50, are often more economical for composite beams in

1.10 SECTION ONE

TABLE 1.5 Relative Material Cost for Beams

Steel

Plate girders

Relativeweight

Relativematerial cost

Rolled beams

Relativeweight

Relativematerial cost

A36 1.000 1.00 1.000 1.00

A572 grade 42 0.927 1.01 0.903 0.98

A572 grade 50 0.848 0.95 0.805 0.91

A588 grade A 0.848 1.04 0.805 0.99

A852 0.775 1.18

A514 grade B 0.600 1.24

the floors of buildings. Also, A588 steel is often preferred for bridge members in view ofits greater durability.

Columns. The relative material cost for two columns of different steels designed to carrythe same load may be expressed as

C F p F /p2 c1 2 c1 1� � (1.7)C F p F /p1 c2 1 c2 2

where Fc1 and Fc2 are the column buckling stresses for the two members. This relationshipis similar to that given for tension members, except that buckling stress is used instead ofyield stress in computing the relative price-strength ratios. Buckling stresses can be calculatedfrom basic column-strength criteria. (T. Y. Galambos, Structural Stability Research CouncilGuide to Design Criteria for Metal Structures, John Wiley & Sons, Inc., New York.) Ingeneral, the buckling stress is considered equal to the yield stress at a slenderness ratio L /rof zero and decreases to the classical Euler value with increasing L /r.

Relative price-strength ratios for A572 grade 50 and other steels, at L /r values from zeroto 120 are shown graphically in Fig. 1.3. As before, A36 steel is the base. Therefore, ratiosless than 1.00 indicate a material cost lower than that of A36 steel. The figure shows thatfor L /r from zero to about 100, A572 grade 50 steel is more economical than A36 steel.Thus the former is frequently used for columns in building construction, particularly in thelower stories, where slenderness ratios are smaller than in the upper stories.

1.4 STEEL SHEET AND STRIP FOR STRUCTURAL APPLICATIONS

Steel sheet and strip are used for many structural applications, including cold-formed mem-bers in building construction and the stressed skin of transportation equipment. Mechanicalproperties of several of the more frequently used sheet steels are presented in Table 1.6.

ASTM A570 covers seven strength grades of uncoated, hot-rolled, carbon-steel sheetsand strip intended for structural use.

A606 covers high-strength, low-alloy, hot- and cold-rolled steel sheet and strip with en-hanced corrosion resistance. This material is intended for structural or miscellaneous useswhere weight savings or high durability are important. It is available, in cut lengths or coils,in either type 2 or type 4, with atmospheric corrosion resistance approximately two or fourtimes, respectively, that of plain carbon steel.


FIGURE 1.3 Curves show for several structural steels the variation of relative price-strengthratios, A36 steel being taken as unity, with slenderness ratios of compression members.

A607, available in six strength levels, covers high-strength, low-alloy columbium or va-nadium, or both, hot- and cold-rolled steel sheet and strip. The material may be in eithercut lengths or coils. It is intended for structural or miscellaneous uses where greater strengthand weight savings are important. A607 is available in two classes, each with six similarstrength levels, but class 2 offers better formability and weldability than class 1. Withoutaddition of copper, these steels are equivalent in resistance to atmospheric corrosion to plaincarbon steel. With copper, however, resistance is twice that of plain carbon steel.

A611 covers cold-rolled carbon sheet steel in coils and cut lengths. Four grades provideyield stress levels from 25 to 40 ksi. Also available is Grade E, which is a full-hard productwith a minimum yield stress of 80 ksi but no specified minimum elongation.

A653 covers steel sheet, zinc coated (galvanized) or zinc-iron alloy coated (galvannealed)by the hot dip process in coils and cut lengths. Included are several grades of structural steel(SS) and high-strength low-alloy steel (HSLAS) with a yield stress of 33 to 80 ksi. HSLASsheets are available as Type A, for applications where improved formability is important,and Type B for even better formability. The metallic coating is available in a wide range ofcoating weights, which provide excellent corrosion protection in many applications.

A715 provides for HSLAS, hot and cold-rolled, with improved formability over A606 anA607 steels. Yield stresses included range from 50 to 80 ksi.

A792 covers sheet in coils and cut lengths coated with aluminum-zinc alloy by the hotdip process. The coating is available in three coating weights, which provide both corrosionand heat resistance.

1.12 SECTION ONE

TABLE 1.6 Specified Minimum Mechanical Properties for Steel Sheet and Strip for StructuralApplications

ASTMdesignation Grade Type of product

Yieldpoint,

ksi

Tensilestrength,

ksiElongationin 2 in, %*

A570 Hot-rolled30 30 49 2133 33 52 1836 36 53 1740 40 55 1545 45 60 1350 50 65 1155 55 70 9

A606 Hot-rolled, cut length 50 70 22Hot-rolled, coils 45 65 22Cold-rolled 45 65 22

A607 Hot- or cold-rolled45 45 60† 25–2250 50 65† 22–2055 55 70† 20–1860 60 75† 18–1665 65 80† 16–1470 70 85† 14–12

A611 Cold-rolledA 25 42 26B 30 45 24C 33 48 22D 40 52 20

A653** Galvanized or galvannealedSS 33 33 45 20SS 37 37 52 18SS 40 40 55 16

SS 50, class 1 50 65 12SS 50, class 2 50 70 12

HSLAS 50 50 60 20–22HSLAS 50 60 70 16–18HSLAS 50 70 80 12–14HSLAS 50 80 90 10–12

A715 Hot- and cold-rolled50 50 60 2260 60 70 1870 70 80 1680 80 90 14

A792 Aluminum-zinc alloy coatedSS 33 33 45 20SS 37 37 52 18SS 40 40 55 16

SS 50A 50 65 12

* Modified for some thicknesses in accordance with the specification. For A607, where two values are given, thefirst is for hot-rolled, the second for cold-rolled steel. For A653, where two values are given, the first is for type Aproduct, the second for type B.

† For class 1 product. Reduce tabulated strengths 5 ksi for class 2.** Also available as A875 with zinc-5% aluminum alloy coating.


TABLE 1.7 Specified Minimum Mechanical Properties of Structural Tubing

ASTMdesignation Product form

Yieldpoint,

ksi

Tensilestrength,

ksiElongationin 2 in, %

A500 ShapedGrade A 39 45 25Grade B 46 58 23Grade C 50 62 21Grade D 36 58 23

A500 RoundGrade A 33 45 25Grade B 42 58 23Grade C 46 62 21Grade D 36 58 23

A501 Round or shaped 36 58 23

A618 Round or shapedGrades Ia, lb, II

Walls �3⁄4 in 50 70 22Walls �3⁄4

to 11⁄2 in 46 67 22Grade III 50 65 20

A847 Round or shaped 50 70 19

1.5 TUBING FOR STRUCTURAL APPLICATIONS

Structural tubing is being used more frequently in modern construction (Art. 6.30). It is oftenpreferred to other steel members when resistance to torsion is required and when a smooth,closed section is aesthetically desirable. In addition, structural tubing often may be the ec-onomical choice for compression members subjected to moderate to light loads. Square andrectangular tubing is manufactured either by cold or hot forming welded or seamless roundtubing in a continuous process. A500 cold-formed carbon-steel tubing (Table 1.7) is producedin four strength grades in each of two product forms, shaped (square or rectangular) orround. A minimum yield point of up to 50 ksi is available for shaped tubes and up to 46ksi for round tubes. A500 grade B and grade C are commonly specified for building con-struction applications and are available from producers and steel service centers.

A501 tubing is a hot-formed carbon-steel product. It provides a yield point equal to thatof A36 steel in tubing having a wall thickness of 1 in or less.

A618 tubing is a hot-formed HSLA product that provides a minimum yield point of upto 50 ksi. The three grades all have enhanced resistance to atmospheric corrosion. GradesIa and Ib can be used in the bare condition for many applications when properly exposedto the atmosphere.

A847 tubing covers cold-formed HSLA tubing and provides a minimum yield point of50 ksi. It also offers enhanced resistance to atmospheric corrosion and, when properly ex-posed, can be used in the bare condition for many applications.

1.6 STEEL CABLE FOR STRUCTURAL APPLICATIONS

Steel cables have been used for many years in bridge construction and are occasionally usedin building construction for the support of roofs and floors. The types of cables used for

1.14 SECTION ONE

TABLE 1.8 Mechanical Properties of Steel Cables

Minimum breaking strength, kip,*of selected cable sizes

Minimum modulus of elasticity, ksi,*for indicated diameter range

Nominaldiameter, in

Zinc-coatedstrand

Zinc-coatedrope

Nominal diameterrange, in

Minimummodulus, ksi

1⁄2 30 23 Prestretched3⁄4 68 52 zinc-coated strand

1 122 91.4 1⁄2 to 29⁄16 24,000

11⁄2 276 208 25⁄8 and over 23,000

2 490 372 Prestretched

3 1076 824 zinc-coated rope

4 1850 1460 3⁄8 to 4 20,000

* Values are for cables with class A zinc coating on all wires. Class B or C can be specified where additionalcorrosion protection is required.

these applications are referred to as bridge strand or bridge rope. In this use, bridge is ageneric term that denotes a specific type of high-quality strand or rope.

A strand is an arrangement of wires laid helically about a center wire to produce asymmetrical section. A rope is a group of strands laid helically around a core composed ofeither a strand or another wire rope. The term cable is often used indiscriminately in referringto wires, strands, or ropes. Strand may be specified under ASTM A586; wire rope, underA603.

During manufacture, the individual wires in bridge strand and rope are generally galva-nized to provide resistance to corrosion. Also, the finished cable is prestretched. In thisprocess, the strand or rope is subjected to a predetermined load of not more than 55% ofthe breaking strength for a sufficient length of time to remove the ‘‘structural stretch’’ causedprimarily by radial and axial adjustment of the wires or strands to the load. Thus, undernormal design loadings, the elongation that occurs is essentially elastic and may be calculatedfrom the elastic-modulus values given in Table 1.8.

Strands and ropes are manufactured from cold-drawn wire and do not have a definiteyield point. Therefore, a working load or design load is determined by dividing the specifiedminimum breaking strength for a specific size by a suitable safety factor. The breakingstrengths for selected sizes of bridge strand and rope are listed in Table 1.8.

1.7 TENSILE PROPERTIES

The tensile properties of steel are generally determined from tension tests on small specimensor coupons in accordance with standard ASTM procedures. The behavior of steels in thesetests is closely related to the behavior of structural-steel members under static loads. Because,for structural steels, the yield points and moduli of elasticity determined in tension andcompression are nearly the same, compression tests are seldom necessary.

Typical tensile stress-strain curves for structural steels are shown in Fig. 1.1. The initialportion of these curves is shown at a magnified scale in Fig. 1.4. Both sets of curves maybe referred to for the following discussion.


FIGURE 1.4 Partial stress-strain curves for structural steels strainedthrough the plastic region into the strain-hardening range. (From R. L.Brockenbrough and B. G. Johnston, USS Steel Design Manual, R. L. Brock-enbrough & Associates, Inc., Pittsburgh, Pa., with permission.)

Strain Ranges. When a steel specimen is subjected to load, an initial elastic range isobserved in which there is no permanent deformation. Thus, if the load is removed, thespecimen returns to its original dimensions. The ratio of stress to strain within the elasticrange is the modulus of elasticity, or Young’s modulus E. Since this modulus is consistentlyabout 29 � 103 ksi for all the structural steels, its value is not usually determined in tensiontests, except in special instances.

The strains beyond the elastic range in the tension test are termed the inelastic range.For as-rolled and high-strength low-alloy (HSLA) steels, this range has two parts. Firstobserved is a plastic range, in which strain increases with no appreciable increase in stress.This is followed by a strain-hardening range, in which strain increase is accompanied bya significant increase in stress. The curves for heat-treated steels, however, do not generallyexhibit a distinct plastic range or a large amount of strain hardening.

The strain at which strain hardening begins (�st) and the rate at which stress increaseswith strain in the strain-hardening range (the strain-hardening modulus Est) have been de-termined for carbon and HSLA steels. The average value of Est is 600 ksi, and the lengthof the yield plateau is 5 to 15 times the yield strain. (T. V. Galambos, ‘‘Properties of Steelfor Use in LRFD,’’ Journal of the Structural Division, American Society of Civil Engineers,Vol. 104, No. ST9, 1978.)

Yield Point, Yield Strength, and Tensile Strength. As illustrated in Fig. 1.4, carbon andHSLA steels usually show an upper and lower yield point. The upper yield point is the valueusually recorded in tension tests and thus is simply termed the yield point.

The heat-treated steels in Fig. 1.4, however, do not show a definite yield point in a tensiontest. For these steels it is necessary to define a yield strength, the stress corresponding to a

1.16 SECTION ONE

specified deviation from perfectly elastic behavior. As illustrated in the figure, yield strengthis usually specified in either of two ways: For steels with a specified value not exceeding80 ksi, yield strength is considered as the stress at which the test specimen reaches a 0.5%extension under load (0.5% EUL) and may still be referred to as the yield point. For higher-strength steels, the yield strength is the stress at which the specimen reaches a strain 0.2%greater than that for perfectly elastic behavior.

Since the amount of inelastic strain that occurs before the yield strength is reached isquite small, yield strength has essentially the same significance in design as yield point.These two terms are sometimes referred to collectively as yield stress.

The maximum stress reached in a tension test is the tensile strength of the steel. Afterthis stress is reached, increasing strains are accompanied by decreasing stresses. Fractureeventually occurs.

Proportional Limit. The proportional limit is the stress corresponding to the first visibledeparture from linear-elastic behavior. This value is determined graphically from the stress-strain curve. Since the departure from elastic action is gradual, the proportional limit dependsgreatly on individual judgment and on the accuracy and sensitivity of the strain-measuringdevices used. The proportional limit has little practical significance and is not usually re-corded in a tension test.

Ductility. This is an important property of structural steels. It allows redistribution ofstresses in continuous members and at points of high local stresses, such as those at holesor other discontinuities.

In a tension test, ductility is measured by percent elongation over a given gage length orpercent reduction of cross-sectional area. The percent elongation is determined by fitting thespecimen together after fracture, noting the change in gage length and dividing the increaseby the original gage length. Similarly, the percent reduction of area is determined from cross-sectional measurements made on the specimen before and after testing.

Both types of ductility measurements are an index of the ability of a material to deformin the inelastic range. There is, however, no generally accepted criterion of minimum ductilityfor various structures.

Poisson’s Ratio. The ratio of transverse to longitudinal strain under load is known as Pois-son’s ratio �. This ratio is about the same for all structural steels—0.30 in the elastic rangeand 0.50 in the plastic range.

True-Stress–True-Strain Curves. In the stress-strain curves shown previously, stress valueswere based on original cross-sectional area, and the strains were based on the original gaugelength. Such curves are sometimes referred to as engineering-type stress-strain curves.However, since the original dimensions change significantly after the initiation of yielding,curves based on instantaneous values of area and gage length are often thought to be ofmore fundamental significance. Such curves are known as true-stress–true-strain curves.A typical curve of this type is shown in Fig. 1.5.

The curve shows that when the decreased area is considered, the true stress actuallyincreases with increase in strain until fracture occurs instead of decreasing after the tensilestrength is reached, as in the engineering stress-strain curve. Also, the value of true strainat fracture is much greater than the engineering strain at fracture (though until yielding beginstrue strain is less than engineering strain).

1.8 PROPERTIES IN SHEAR

The ratio of shear stress to shear strain during initial elastic behavior is the shear modulusG. According to the theory of elasticity, this quantity is related to the modulus of elasticityE and Poisson’s ratio � by


FIGURE 1.5 Curve shows the relationship between true stressand true strain for 50-ksi yield-point HSLA steel.

EG � (1.8)

2(1 � �)

Thus a minimum value of G for structural steels is about 11 � 103 ksi. The yield stress inshear is about 0.57 times the yield stress in tension. The shear strength, or shear stress atfailure in pure shear, varies from two-thirds to three-fourths the tensile strength for thevarious steels. Because of the generally consistent relationship of shear properties to tensileproperties for the structural steels, and because of the difficulty of making accurate sheartests, shear tests are seldom performed.

1.9 HARDNESS TESTS

In the Brinell hardness test, a small spherical ball of specified size is forced into a flat steelspecimen by a known static load. The diameter of the indentation made in the specimen canbe measured by a micrometer microscope. The Brinell hardness number may then becalculated as the ratio of the applied load, in kilograms, to the surface area of the indentation,in square millimeters. In practice, the hardness number can be read directly from tables forgiven indentation measurements.

The Rockwell hardness test is similar in principle to the Brinell test. A spheroconicaldiamond penetrator is sometimes used to form the indentation and the depth of the inden-tation is measured with a built-in, differential depth-measurement device. This measurement,which can be read directly from a dial on the testing device, becomes the Rockwell hardnessnumber.

In either test, the hardness number depends on the load and type of penetrator used;therefore, these should be indicated when listing a hardness number. Other hardness tests,such as the Vickers tests, are also sometimes used. Tables are available that give approximaterelationships between the different hardness numbers determined for a specific material.

Hardness numbers are considered to be related to the tensile strength of steel. Althoughthere is no absolute criterion to convert from hardness numbers to tensile strength, chartsare available that give approximate conversions (see ASTM A370). Because of its simplicity,the hardness test is widely used in manufacturing operations to estimate tensile strength andto check the uniformity of tensile strength in various products.

1.18 SECTION ONE

FIGURE 1.6 Stress-strain diagram (not to scale) illustratingthe effects of strain-hardening steel. (From R. L. Brockenbroughand B. G. Johnston, USS Steel Design Manual, R. L. Brocken-brough & Associates, Inc., Pittsburgh, Pa., with permission.)

1.10 EFFECT OF COLD WORK ON TENSILE PROPERTIES

In the fabrication of structures, steel plates and shapes are often formed at room temperaturesinto desired shapes. These cold-forming operations cause inelastic deformation, since thesteel retains its formed shape. To illustrate the general effects of such deformation on strengthand ductility, the elemental behavior of a carbon-steel tension specimen subjected to plasticdeformation and subsequent tensile reloadings will be discussed. However, the behavior ofactual cold-formed structural members is more complex.

As illustrated in Fig. 1.6, if a steel specimen is unloaded after being stressed into eitherthe plastic or strain-hardening range, the unloading curve follows a path parallel to the elasticportion of the stress-strain curve. Thus a residual strain, or permanent set, remains after theload is removed. If the specimen is promptly reloaded, it will follow the unloading curve tothe stress-strain curve of the virgin (unstrained) material.

If the amount of plastic deformation is less than that required for the onset of strainhardening, the yield stress of the plastically deformed steel is about the same as that of thevirgin material. However, if the amount of plastic deformation is sufficient to cause strainhardening, the yield stress of the steel is larger. In either instance, the tensile strength remainsthe same, but the ductility, measured from the point of reloading, is less. As indicated inFig. 1.6, the decrease in ductility is nearly equal to the amount of inelastic prestrain.

A steel specimen that has been strained into the strain-hardening range, unloaded, andallowed to age for several days at room temperature (or for a much shorter time at a mod-erately elevated temperature) usually shows the behavior indicated in Fig. 1.7 during reload-ing. This phenomenon, known as strain aging, has the effect of increasing yield and tensilestrength while decreasing ductility.


FIGURE 1.7 Effects of strain aging are shown by stress-straindiagram (not to scale). (From R. L. Brockenbrough and B. G.Johnston, USS Steel Design Manual, R. L. Brockenbrough & As-sociates, Inc., Pittsburgh, Pa., with permission.)

Most of the effects of cold work on the strength and ductility of structural steels can beeliminated by thermal treatment, such as stress relieving, normalizing, or annealing. However,such treatment is not often necessary.

(G. E. Dieter, Jr., Mechanical Metallurgy, 3rd ed., McGraw-Hill, Inc., New York.)

1.11 EFFECT OF STRAIN RATE ON TENSILE PROPERTIES

Tensile properties of structural steels are usually determined at relatively slow strain rates toobtain information appropriate for designing structures subjected to static loads. In the designof structures subjected to high loading rates, such as those caused by impact loads, however,it may be necessary to consider the variation in tensile properties with strain rate.

Figure 1.8 shows the results of rapid tension tests conducted on a carbon steel, two HSLAsteels, and a constructional alloy steel. The tests were conducted at three strain rates and atthree temperatures to evaluate the interrelated effect of these variables on the strength of thesteels. The values shown for the slowest and the intermediate strain rates on the room-temperature curves reflect the usual room-temperature yield stress and tensile strength, re-spectively. (In determination of yield stress, ASTM E8 allows a maximum strain rate of 1⁄16

in per in per mm, or 1.04 � 10�3 in per in per sec. In determination of tensile strength, E8allows a maximum strain rate of 0.5 in per in per mm, or 8.33 � 10�3 in per in per sec.)

The curves in Fig. 1.8a and b show that the tensile strength and 0.2% offset yield strengthof all the steels increase as the strain rate increases at �50�F and at room temperature. Thegreater increase in tensile strength is about 15%, for A514 steel, whereas the greatest increasein yield strength is about 48%, for A515 carbon steel. However, Fig. 1.8c shows that at600�F, increasing the strain rate has a relatively small influence on the yield strength. But afaster strain rate causes a slight decrease in the tensile strength of most of the steels.

1.20 SECTION ONE

FIGURE 1.8 Effects of strain rate on yield and tensile strengths of structural steels at low, normal,and elevated temperatures. (From R. L. Brockenbrough and B. G. Johnston, USS Steel DesignManual, R. L. Brockenbrough & Associates, Inc., Pittsburgh, Pa., with permission.)

Ductility of structural steels, as measured by elongation or reduction of area, tends todecrease with strain rate. Other tests have shown that modulus of elasticity and Poisson’sratio do not vary significantly with strain rate.

1.12 EFFECT OF ELEVATED TEMPERATURES ON TENSILEPROPERTIES

The behavior of structural steels subjected to short-time loadings at elevated temperatures isusually determined from short-time tension tests. In general, the stress-strain curve becomesmore rounded and the yield strength and tensile strength are reduced as temperatures areincreased. The ratios of the elevated-temperature value to room-temperature value of yieldand tensile strengths of several structural steels are shown in Fig. 1.9a and b, respectively.

Modulus of elasticity decreases with increasing temperature, as shown in Fig. 1.9c. Therelationship shown is nearly the same for all structural steels. The variation in shear moduluswith temperature is similar to that shown for the modulus of elasticity. But Poisson’s ratiodoes not vary over this temperature range.

The following expressions for elevated-temperature property ratios, which were derivedby fitting curves to short-time data, have proven useful in analytical modeling (R. L. Brock-enbrough, ‘‘Theoretical Stresses and Strains from Heat Curving,’’ Journal of the StructuralDivision, American Society of Civil Engineers, Vol. 96, No. ST7, 1970):

1.21

FIGURE 1.9 Effect of temperature on (a) yield strengths, (b) tensile strengths, and(c) modulus of elasticity of structural steels. (From R. L. Brockenbrough and B. G.Johnston, USS Steel Design Manual, R. L. Brockenbrough & Associates, Inc., Pitts-burgh, Pa., with permission.)

1.22 SECTION ONE

T � 100F /F� � 1 � 100�F � T � 800�F (1.9)y y 5833

2 �6F /F� � (�720,000 � 4200 � 2.75T )10 800�F � T � 1200�F (1.10)y y

T � 100E /E� � 1 � 100�F � T � 700�F (1.11)

5000

2 �6E /E� � (500,000 � 1333T � 1.111T )10 700�F � T � 1200�F (1.12)

�6� � (6.1 � 0.0019T )10 100�F � T � 1200�F (1.13)

In these equations Fy / and E /E� are the ratios of elevated-temperature to room-temperatureF�yyield strength and modulus of elasticity, respectively, � is the coefficient of thermal expansionper degree Fahrenheit, and T is the temperature in degrees Fahrenheit.

Ductility of structural steels, as indicated by elongation and reduction-of-area values,decreases with increasing temperature until a minimum value is reached. Thereafter, ductilityincreases to a value much greater than that at room temperature. The exact effect dependson the type and thickness of steel. The initial decrease in ductility is caused by strain agingand is most pronounced in the temperature range of 300 to 700�F. Strain aging also accountsfor the increase in tensile strength in this temperature range shown for two of the steels inFig. 1.9b.

Under long-time loadings at elevated temperatures, the effects of creep must be consid-ered. When a load is applied to a specimen at an elevated temperature, the specimen deformsrapidly at first but then continues to deform, or creep, at a much slower rate. A schematiccreep curve for a steel subjected to a constant tensile load and at a constant elevated tem-perature is shown in Fig. 1.10. The initial elongation occurs almost instantaneously and isfollowed by three stages. In stage 1 elongation increases at a decreasing rate. In stage 2,elongation increases at a nearly constant rate. And in stage 3, elongation increases at anincreasing rate. The failure, or creep-rupture, load is less than the load that would causefailure at that temperature in a short-time loading test.

Table 1.9 indicates typical creep and rupture data for a carbon steel, an HSLA steel, anda constructional alloy steel. The table gives the stress that will cause a given amount ofcreep in a given time at a particular temperature.

For special elevated-temperature applications in which structural steels do not provideadequate properties, special alloy and stainless steels with excellent high-temperature prop-erties are available.

1.13 FATIGUE

A structural member subjected to cyclic loadings may eventually fail through initiation andpropagation of cracks. This phenomenon is called fatigue and can occur at stress levelsconsiderably below the yield stress.

Extensive research programs conducted to determine the fatigue strength of structuralmembers and connections have provided information on the factors affecting this property.These programs included studies of large-scale girder specimens with flange-to-web filletwelds, flange cover plates, stiffeners, and other attachments. The studies showed that thestress range (algebraic difference between maximum and minimum stress) and notch se-verity of details are the most important factors. Yield point of the steel had little effect. Theknowledge developed from these programs has been incorporated into specifications of theAmerican Institute of Steel Construction, American Association of State Highway and Trans-portation Officials, and the American Railway Engineering and Maintenance-of-Way Asso-ciation, which offer detailed provisions for fatigue design.


FIGURE 1.10 Creep curve for structural steel in tension (sche-matic). (From R. L. Brockenbrough and B. G. Johnston, USS SteelDesign Manual, R. L. Brockenbrough & Associates, Inc., Pitts-burgh, Pa., with permission.)

1.14 BRITTLE FRACTURE

Under sufficiently adverse combinations of tensile stress, temperature, loading rate, geometricdiscontinuity (notch), and restraint, a steel member may experience a brittle fracture. Allthese factors need not be present. In general, a brittle fracture is a failure that occurs bycleavage with little indication of plastic deformation. In contrast, a ductile fracture occursmainly by shear, usually preceded by considerable plastic deformation.

Design against brittle fracture requires selection of the proper grade of steel for the ap-plication and avoiding notchlike defects in both design and fabrication. An awareness of thephenomenon is important so that steps can be taken to minimize the possibility of thisundesirable, usually catastrophic failure mode.

An empirical approach and an analytical approach directed toward selection and evalua-tion of steels to resist brittle fracture are outlined below. These methods are actually com-plementary and are frequently used together in evaluating material and fabrication require-ments.

Charpy V-Notch Test. Many tests have been developed to rate steels on their relative re-sistance to brittle fracture. One of the most commonly used tests is the Charpy V-notch test,which specifically evaluates notch toughness, that is, the resistance to fracture in the presenceof a notch. In this test, a small square bar with a specified-size V-shaped notch at its mid-length (type A impact-test specimen of ASTM A370) is simply supported at its ends as abeam and fractured by a blow from a swinging pendulum. The amount of energy requiredto fracture the specimen or the appearance of the fracture surface is determined over a rangeof temperatures. The appearance of the fracture surface is usually expressed as the percentageof the surface that appears to have fractured by shear.

1.24 SECTION ONE

TABLE 1.9 Typical Creep Rates and Rupture Stresses for Structural Steels at Various Temperatures

Testtemperature,

�F

Stress, ksi, for creep rate of

0.0001% per hr* 0.00001% per hr†

Stress, ksi for rupture in

1000 hours 10,000 hours 100,000 hours

A36 steel

800 21.4 13.8 38.0 24.8 16.0

900 9.9 6.0 18.5 12.4 8.2

1000 4.6 2.6 9.5 6.3 4.2

A588 grade A steel†

800 34.6 29.2 44.1 35.7 28.9

900 20.3 16.3 28.6 22.2 17.3

1000 11.4 8.6 17.1 12.0 8.3

1200 1.7 1.0 3.8 2.0 1.0

A514 grade F steel†

700 — — 101.0 99.0 97.0

800 81.0 74.0 86.0 81.0 77.0

* Equivalent to 1% in 10,000 hours.† Equivalent to 1% in 100,000 hours.‡ Not recommended for use where temperatures exceed 800�F.

FIGURE 1.11 Transition curves from Charpy-V notch impact tests. (a) Variation of percent shearfracture with temperature. (b) Variation of absorbed energy with temperature.

A shear fracture is indicated by a dull or fibrous appearance. A shiny or crystallineappearance is associated with a cleavage fracture.

The data obtained from a Charpy test are used to plot curves, such as those in Fig. 1.11,of energy or percentage of shear fracture as a function of temperature. The temperature nearthe bottom of the energy-temperature curve, at which a selected low value of energy isabsorbed, often 15 ft � lb, is called the ductility transition temperature or the 15-ft � lb


FIGURE 1.12 Fracture mechanics analysis for brittle fracture. (a) Sharp crack in a stressed infiniteplate. (b) Disk-shaped crack in an infinite body. (c) Relation of fracture toughness to thickness.

transition temperature. The temperature at which the percentage of shear fracture decreasesto 50% is often called the fracture-appearance transition temperature. These transitiontemperatures serve as a rating of the resistance of different steels to brittle fracture. Thelower the transition temperature, the greater is the notch toughness.

Of the steels in Table 1.1, A36 steel generally has about the highest transition temperature.Since this steel has an excellent service record in a variety of structural applications, itappears likely that any of the structural steels, when designed and fabricated in an appropriatemanner, could be used for similar applications with little likelihood of brittle fracture. Nev-ertheless, it is important to avoid unusual temperature, notch, and stress conditions to min-imize susceptibility to brittle fracture.

In applications where notch toughness is considered important, the minimum CharpyV-notch value and test temperature should be specified, because there may be considerablevariation in toughness within any given product designation unless specifically produced tominimum requirements. The test temperature may be specified higher than the lowest op-erating temperature to compensate for a lower rate of loading in the anticipated application.(See Art. 1.1.5.)

It should be noted that as the thickness of members increases, the inherent restraintincreases and tends to inhibit ductile behavior. Thus special precautions or greater toughness,or both, is required for tension or flexural members comprised of thick material. (See Art.1.17.)

Fracture-Mechanics Analysis. Fracture mechanics offers a more direct approach for pre-diction of crack propagation. For this analysis, it is assumed that a crack, which may bedefined as a flat, internal defect, is always present in a stressed body. By linear-elastic stressanalysis and laboratory tests on a precracked specimen, the defect size is related to theapplied stress that will cause crack propagation and brittle fracture, as outlined below.

Near the tip of a crack, the stress component ƒ perpendicular to the plane of the crack(Fig. 1.12a) can be expressed as

KIƒ � (1.14)�2�r

where r is distance from tip of crack and KI is a stress-intensity factor related to geometry

1.26 SECTION ONE

of crack and to applied loading. The factor KI can be determined from elastic theory forgiven crack geometries and loading conditions. For example, for a through-thickness crackof length 2a in an infinite plate under uniform stress (Fig. 1.12a),

K � ƒ ��a (1.15)I a

where ƒa is the nominal applied stress. For a disk-shaped crack of diameter 2a embedded inan infinite body (Fig. 1.12b), the relationship is

aK � 2ƒ (1.16)I a ��

If a specimen with a crack of known geometry is loaded until the crack propagates rapidlyand causes failure, the value of KI at that stress level can be calculated from the derivedexpression. This value is termed the fracture toughness Kc.

A precracked tension or bend-type specimen is usually used for such tests. As the thick-ness of the specimen increases and the stress condition changes from plane stress to planestrain, the fracture toughness decreases to a minimum value, as illustrated in Fig. 1.12c. Thisvalue of plane-strain fracture toughness designated KIc, may be regarded as a fundamentalmaterial property.

Thus, if KIc is substituted for KI, for example, in Eq. (1.15) or (1.16) a numerical rela-tionship is obtained between the crack geometry and the applied stress that will cause frac-ture. With this relationship established, brittle fracture may be avoided by determining themaximum-size crack present in the body and maintaining the applied stress below the cor-responding level. The tests must be conducted at or correlated with temperatures and strainrates appropriate for the application, because fracture toughness decreases with temperatureand loading rate. Correlations have been made to enable fracture toughness values to beestimated from the results of Charpy V-notch tests.

Fracture-mechanics analysis has proven quite useful, particularly in critical applications.Fracture-control plans can be established with suitable inspection intervals to ensure thatimperfections, such as fatigue cracks do not grow to critical size.

(J. M. Barsom and S. T. Rolfe, Fracture and Fatigue Control in Structures; Applicationsof Fracture Mechanics, Prentice-Hall, Inc. Englewood Cliffs, N.J.)

1.15 RESIDUAL STRESSES

Stresses that remain in structural members after rolling or fabrication are known as residualstresses. The magnitude of the stresses is usually determined by removing longitudinal sec-tions and measuring the strain that results. Only the longitudinal stresses are usually mea-sured. To meet equilibrium conditions, the axial force and moment obtained by integratingthese residual stresses over any cross section of the member must be zero.

In a hot-rolled structural shape, the residual stresses result from unequal cooling ratesafter rolling. For example, in a wide-flange beam, the center of the flange cools more slowlyand develops tensile residual stresses that are balanced by compressive stresses elsewhereon the cross section (Fig. 1.13a). In a welded member, tensile residual stresses develop nearthe weld and compressive stresses elsewhere provide equilibrium, as shown for the weldedbox section in Fig. 1.13b.

For plates with rolled edges (UM plates), the plate edges have compressive residualstresses (Fig. 1.13c). However, the edges of flame-cut plates have tensile residual stresses(Fig. 1.13d ). In a welded I-shaped member, the stress condition in the edges of flangesbefore welding is reflected in the final residual stresses (Fig. 1.13e). Although not shown inFig. 1.13, the residual stresses at the edges of sheared-edge plates vary through the plate


FIGURE 1.13 Typical residual-stress distributions (� indicates tension and � com-pression).

thickness. Tensile stresses are present on one surface and compressive stresses on the op-posite surface.

The residual-stress distributions mentioned above are usually relatively constant along thelength of the member. However, residual stresses also may occur at particular locations in amember, because of localized plastic flow from fabrication operations, such as cold straight-ening or heat straightening.

When loads are applied to structural members, the presence of residual stresses usuallycauses some premature inelastic action; that is, yielding occurs in localized portions beforethe nominal stress reaches the yield point. Because of the ductility of steel, the effect onstrength of tension members is not usually significant, but excessive tensile residual stresses,in combination with other conditions, can cause fracture. In compression members, residualstresses decrease the buckling load from that of an ideal or perfect member. However, currentdesign criteria in general use for compression members account for the influence of residualstress.

In bending members that have residual stresses, a small inelastic deflection of insignificantmagnitude may occur with the first application of load. However, under subsequent loads ofthe same magnitude, the behavior is elastic. Furthermore, in ‘‘compact’’ bending members,the presence of residual stresses has no effect on the ultimate moment (plastic moment).Consequently, in the design of statically loaded members, it is not usually necessary toconsider residual stresses.

1.28 SECTION ONE

1.16 LAMELLAR TEARING

In a structural steel member subjected to tension, elongation and reduction of area in sectionsnormal to the stress are usually much lower in the through-thickness direction than in theplanar direction. This inherent directionality is of small consequence in many applications,but it does become important in design and fabrication of structures with highly restrainedjoints because of the possibility of lamellar tearing. This is a cracking phenomenon thatstarts underneath the surface of steel plates as a result of excessive through-thickness strain,usually associated with shrinkage of weld metal in highly restrained joints. The tear has asteplike appearance consisting of a series of terraces parallel to the surface. The crackingmay remain completely below the surface or may emerge at the edges of plates or shapesor at weld toes.

Careful selection of weld details, filler metal, and welding procedure can restrict lamellartearing in heavy welded constructions, particularly in joints with thick plates and heavystructural shapes. Also, when required, structural steels can be produced by special processes,generally with low sulfur content and inclusion control, to enhance through-thickness duc-tility.

The most widely accepted method of measuring the susceptibility of a material to lamellartearing is the tension test on a round specimen, in which is observed the reduction in areaof a section oriented perpendicular to the rolled surface. The reduction required for a givenapplication depends on the specific details involved. The specifications to which a particularsteel can be produced are subject to negotiations with steel producers.

(R. L. Brockenbrough, Chap. 1.2 in Constructional Steel Design—An International Guide,R. Bjorhovde et al., eds., Elsevier Science Publishers, Ltd., New York.)

1.17 WELDED SPLICES IN HEAVY SECTIONS

Shrinkage during solidification of large welds in structural steel members causes, in adjacentrestrained metal, strains that can exceed the yield-point strain. In thick material, triaxialstresses may develop because there is restraint in the thickness direction as well as in planardirections. Such conditions inhibit the ability of a steel to act in a ductile manner and increasethe possibility of brittle fracture. Therefore, for members subject to primary tensile stressesdue to axial tension or flexure in buildings, the American Institute of Steel Construction(AISC) specifications for structural steel buildings impose special requirements for weldedsplicing of either group 4 or group 5 rolled shapes or of shapes built up by welding platesmore than 2 in thick. The specifications include requirements for notch toughness, removalof weld tabs and backing bars (welds ground smooth), generous-sized weld-access holes,preheating for thermal cutting, and grinding and inspecting cut edges. Even for primarycompression members, the same precautions should be taken for sizing weld access holes,preheating, grinding, and inspection.

Most heavy wide-flange shapes and tees cut from these shapes have regions where thesteel has low toughness, particularly at flange-web intersections. These low-toughness regionsoccur because of the slower cooling there and, because of the geometry, the lower rollingpressure applied there during production. Hence, to ensure ductility and avoid brittle failure,bolted splices should be considered as an alternative to welding.

(‘‘AISC Specification for Structural Steel Buildings—Allowable Stress Design and PlasticDesign’’ and ‘‘Load and Resistance Factor Design Specification for Structural Steel Build-ings,’’ American Institute of Steel Construction; R. L. Brockenbrough, Sec. 9, in StandardHandbook for Civil Engineers, 4th ed., McGraw-Hill, Inc., New York.)


1.18 k-AREA CRACKING

Wide flange sections are typically straightened as part of the mill production process. Oftena rotary straightening process is used, although some heavier members may be straightenedin a gag press. Some reports in recent years have indicated a potential for crack initiationat or near connections in the ‘‘k’’ area of wide flange sections that have been rotary straight-ened. The k area is the region extending from approximately the mid-point of the web-to-flange fillet, into the web for a distance approximately 1 to 11⁄2 in beyond the point oftangency. Apparently, in some cases, this limited region had a reduced notch toughness dueto cold working and strain hardening. Most of the incidents reported occurred at highlyrestrained joints with welds in the k area. However, the number of examples reported hasbeen limited and these have occurred during construction or laboratory tests, with no evi-dence of difficulties with steel members in service.

Research sponsored by AISC is underway to define the extent of the problem and makeappropriate recommendations. Until further information is available, AISC has issued thefollowing recommendations concerning fabrication and design practices for rolled wideflange shapes:

• Welds should be stopped short of the ‘‘k’’ area for transverse stiffeners (continuity plates).

• For continuity plates, fillet welds and/or partial joint penetration welds, proportioned totransfer the calculated stresses to the column web, should be considered instead of com-plete joint penetration welds. Weld volume should be minimized.

• Residual stresses in highly restrained joints may be decreased by increased preheat andproper weld sequencing.

• Magnetic particle or dye penetrant inspection should be considered for weld areas in ornear the k area of highly restrained connections after the final welding has completelycooled.

• When possible, eliminate the need for column web doubler plates by increasing columnsize.

Good fabrication and quality control practices, such as inspection for cracks and gougesat flame-cut access holes or copes, should continue to be followed and any defects repairedand ground smooth. All structural wide flange members for normal service use in buildingconstruction should continue to be designed per AISC Specifications and the material fur-nished per ASTM standards.’’

(AISC Advisory Statement, Modern Steel Construction, February 1997.)

1.19 VARIATIONS IN MECHANICAL PROPERTIES

Tensile properties of structural steel may vary from specified minimum values. Product spec-ifications generally require that properties of the material ‘‘as represented by the test speci-men’’ meet certain values. With some exceptions, ASTM specifications dictate a test fre-quency for structural-grade steels of only two tests per heat (in each strength level produced,if applicable) and more frequent testing for pressure-vessel grades. If the heats are very large,the test specimens qualify a considerable amount of product. As a result, there is a possibilitythat properties at locations other than those from which the specimens were taken will bedifferent from those specified.

For plates, a test specimen is required by ASTM A6 to be taken from a corner. If theplates are wider than 24 in, the longitudinal axis of the specimen should be oriented trans-

1.30 SECTION ONE

versely to the final direction in which the plates were rolled. For other products, however,the longitudinal axis of the specimen should be parallel to the final direction of rolling.

For structural shapes with a flange width of 6 in or more, test specimens should beselected from a point in the flange as near as practicable to 2⁄3 the distance from the flangecenterline to the flange toe. Prior to 1997–1998, the specimens were taken from the web.

An extensive study commissioned by the American Iron and Steel Institute (AISI) com-pared yield points at various sample locations with the official product test. The studiesindicated that the average difference at the check locations was �0.7 ksi. For the top andbottom flanges, at either end of beams, the average difference at check locations was �2.6ksi.

Although the test value at a given location may be less than that obtained in the officialtest, the difference is offset to the extent that the value from the official test exceeds thespecified minimum value. For example, a statistical study made to develop criteria for loadand resistance factor design showed that the mean yield points exceeded the specified min-imum yield point Fy (specimen located in web) as indicated below and with the indicatedcoefficient of variation (COV).

Flanges of rolled shapes 1.05Fy, COV � 0.10Webs of rolled shapes 1.10Fy, COV � 0.11Plates 1.10Fy, COV � 0.11

Also, these values incorporate an adjustment to the lower ‘‘static’’ yield points.For similar reasons, the notch toughness can be expected to vary throughout a product.(R. L. Brockenbrough, Chap. 1.2, in Constructional Steel Design—An International

Guide, R. Bjorhovde, ed., Elsevier Science Publishers, Ltd., New York.)

1.20 CHANGES IN CARBON STEELS ON HEATING ANDCOOLING*

As pointed out in Art. 1.12, heating changes the tensile properties of steels. Actually, heatingchanges many steel properties. Often, the primary reason for such changes is a change instructure brought about by heat. Some of these structural changes can be explained with theaid of an iron-carbon equilibrium diagram (Fig. 1.14).

The diagram maps out the constituents of carbon steels at various temperatures as carboncontent ranges from 0 to 5%. Other elements are assumed to be present only as impurities,in negligible amounts.

If a steel with less than 2% carbon is very slowly cooled from the liquid state, a solidsolution of carbon in gamma iron will result. This is called austenite. (Gamma iron is apure iron whose crystalline structure is face-centered cubic.)

If the carbon content is about 0.8%, the carbon remains in solution as the austenite slowlycools, until the A1 temperature (1340�F) is reached. Below this temperature, the austenitetransforms to the eutectoid pearlite. This is a mixture of ferrite and cementite (iron carbide,Fe3C). Pearlite, under a microscope, has a characteristic platelike, or lamellar, structure withan iridescent appearance, from which it derives its name.

If the carbon content is less than 0.8%, as is the case with structural steels, coolingaustenite below the A3 temperature line causes transformation of some of the austenite toferrite. (This is a pure iron, also called alpha iron, whose crystalline structure is body-centered cubic.) Still further cooling to below the A1 line causes the remaining austenite to

*Articles 1.20 through 1.28 adapted from previous edition written by Frederick S. Merritt, Consulting Engineer,West Palm Beach, Florida.


FIGURE 1.14 Iron-carbon equilibrium diagram.

transform to pearlite. Thus, as indicated in Fig. 1.14, low-carbon steels are hypoeutectoidsteels, mixtures of ferrite and pearlite.

Ferrite is very ductile but has low tensile strength. Hence carbon steels get their highstrengths from the pearlite present or, more specifically, from the cementite in the pearlite.

The iron-carbon equilibrium diagram shows only the constituents produced by slow cool-ing. At high cooling rates, however, equilibrium cannot be maintained. Transformation tem-peratures are lowered, and steels with microstructures other than pearlitic may result. Prop-erties of such steels differ from those of the pearlitic steels. Heat treatments of steels arebased on these temperature effects.

If a low-carbon austenite is rapidly cooled below about 1300�F, the austenite will trans-form at constant temperature into steels with one of four general classes of microstructure:

Pearlite, or lamellar, microstructure results from transformations in the range 1300 to1000�F. The lower the temperature, the closer is the spacing of the platelike elements. Asthe spacing becomes smaller, the harder and tougher the steels become. Steels such as A36,A572, and A588 have a mixture of a soft ferrite matrix and a hard pearlite.

Bainite forms in transformations below about 1000�F and above about 450�F. It has anacicular, or needlelike, microstructure. At the higher temperatures, bainite may be softer thanthe pearlitic steels. However, as the transformation temperature is decreased, hardness andtoughness increase.

Martensite starts to form at a temperature below about 500�F, called the Ms temperature.The transformation differs from those for pearlitic and bainitic steels in that it is not time-dependent. Martensite occurs almost instantly during rapid cooling, and the percentage ofaustenite transformed to martensite depends only on the temperature to which the steel iscooled. For complete conversion to martensite, cooling must extend below the Mƒ temper-ature, which may be 200�F or less. Like bainite, martensite has an acicular microstructure,but martensite is harder and more brittle than pearlitic and bainitic steels. Its hardness varieswith carbon content and to some extent with cooling rate. For some applications, such asthose where wear resistance is important, the high hardness of martensite is desirable, despitebrittleness. Generally, however, martensite is used to obtain tempered martensite, which hassuperior properties.

Tempered martensite is formed when martensite is reheated to a subcritical temperatureafter quenching. The tempering precipitates and coagulates carbides. Hence the microstruc-ture consists of carbide particles, often spheroidal in shape, dispersed in a ferrite matrix. The

1.32 SECTION ONE

result is a loss in hardness but a considerable improvement in ductility and toughness. Theheat-treated carbon and HSLA steels and quenched and tempered constructional steels dis-cussed in Art. 1.1 are low-carbon martensitic steels.

(Z. D. Jastrzebski, Nature and Properties of Engineering Materials, John Wiley & Sons,Inc., New York.)

1.21 EFFECTS OF GRAIN SIZE

As indicated in Fig. 1.14, when a low-carbon steel is heated above the A1 temperature line,austenite, a solid solution of carbon in gamma iron, begins to appear in the ferrite matrix.Each island of austenite grows until it intersects its neighbor. With further increase in tem-perature, these grains grow larger. The final grain size depends on the temperature above theA3 line to which the metal is heated. When the steel cools, the relative coarseness of thegrains passes to the ferrite-plus-pearlite phase.

At rolling and forging temperatures, therefore, many steels grow coarse grains. Hot work-ing, however, refines the grain size. The temperature at the final stage of the hot-workingprocess determines the final grain size. When the finishing temperature is relatively high,the grains may be rather coarse when the steel is air-cooled. In that case, the grain size canbe reduced if the steel is normalized (reheated to just above the A3 line and again air-cooled).(See Art. 1.22.)

Fine grains improve many properties of steels. Other factors being the same, steels withfiner grain size have better notch toughness because of lower transition temperatures (seeArt. 1.14) than coarser-grained steels. Also, decreasing grain size improves bendability andductility. Furthermore fine grain size in quenched and tempered steel improves yield strength.And there is less distortion, less quench cracking, and lower internal stress in heat-treatedproducts.

On the other hand, for some applications, coarse-grained steels are desirable. They permitdeeper hardening. If the steels should be used in elevated-temperature service, they offerhigher load-carrying capacity and higher creep strength than fine-grained steels.

Austenitic-grain growth may be inhibited by carbides that dissolve slowly or remainundissolved in the austenite or by a suitable dispersion of nonmetallic inclusions. Steelsproduced this way are called fine-grained. Steels not made with grain-growth inhibitors arecalled coarse-grained.

When heated above the critical temperature, 1340�F, grains in coarse-grained steels growgradually. The grains in fine-grained steels grow only slightly, if at all, until a certain tem-perature, the coarsening temperature, is reached. Above this, abrupt coarsening occurs. Theresulting grain size may be larger than that of coarse-grained steel at the same temperature.Note further that either fine-grained or coarse-grained steels can be heat-treated to be eitherfine-grained or coarse-grained (see Art. 1.22).

The usual method of making fine-grained steels involves controlled aluminum deoxidation(see also Art. 1.24). The inhibiting agent in such steels may be a submicroscopic dispersionof aluminum nitride or aluminum oxide.

(W. T. Lankford, Jr., ed., The Making, Shaping and Treating of Steel, Association of Ironand Steel Engineers, Pittsburgh, Pa.)

1.22 ANNEALING AND NORMALIZING

Structural steels may be annealed to relieve stresses induced by cold or hot working. Some-times, also, annealing is used to soften metal to improve its formability or machinability.


Annealing involves austenitizing the steel by heating it above the A3 temperature line inFig. 1.14, then cooling it slowly, usually in a furnace. This treatment improves ductility butdecreases tensile strength and yield point. As a result, further heat treatment may be nec-essary to improve these properties.

Structural steels may be normalized to refine grain size. As pointed out in Art. 1.21, grainsize depends on the finishing temperature in hot rolling.

Normalizing consists of heating the steel above the A3 temperature line, then cooling themetal in still air. Thus the rate of cooling is more rapid than in annealing. Usual practice isto normalize from 100 to 150�F above the critical temperature. Higher temperatures coarsenthe grains.

Normalizing tends to improve notch toughness by lowering ductility and fracture transi-tion temperatures. Thick plates benefit more from this treatment than thin plates. Requiringfewer roller passes, thick plates have a higher finishing temperature and cool slower thanthin plates, thus have a more adverse grain structure. Hence the improvement from normal-izing is greater for thick plates.

1.23 EFFECTS OF CHEMISTRY ON STEEL PROPERTIES

Chemical composition determines many characteristics of steels important in constructionapplications. Some of the chemicals present in commercial steels are a consequence of thesteelmaking process. Other chemicals may be added deliberately by the producers to achievespecific objectives. Specifications therefore usually require producers to report the chemicalcomposition of the steels.

During the pouring of a heat of steel, producers take samples of the molten steel forchemical analysis. These heat analyses are usually supplemented by product analyses takenfrom drillings or millings of blooms, billets, or finished products. ASTM specifications con-tain maximum and minimum limits on chemicals reported in the heat and product analyses,which may differ slightly.

Principal effects of the elements more commonly found in carbon and low-alloy steelsare discussed below. Bear in mind, however, that the effects of two or more of these chem-icals when used in combination may differ from those when each alone is present. Note alsothat variations in chemical composition to obtain specific combinations of properties in asteel usually increase cost, because it becomes more expensive to make, roll, and fabricate.

Carbon is the principal strengthening element in carbon and low-alloy steels. In general,each 0.01% increase in carbon content increases the yield point about 0.5 ksi. This, however,is accompanied by increase in hardness and reduction in ductility, notch toughness, andweldability, raising of the transition temperatures, and greater susceptibility to aging. Hencelimits on carbon content of structural steels are desirable. Generally, the maximum permittedin structural steels is 0.30% or less, depending on the other chemicals present and the weld-ability and notch toughness desired.

Aluminum, when added to silicon-killed steel, lowers the transition temperature andincreases notch toughness. If sufficient aluminum is used, up to about 0.20%, it reduces thetransition temperature even when silicon is not present. However, the larger additions ofaluminum make it difficult to obtain desired finishes on rolled plate. Drastic deoxidation ofmolten steels with aluminum or aluminum and titanium, in either the steelmaking furnaceor the ladle, can prevent the spontaneous increase in hardness at room temperature calledaging. Also, aluminum restricts grain growth during heat treatment and promotes surfacehardness by nitriding.

Boron in small quantities increases hardenability of steels. It is used for this purpose inquenched and tempered low-carbon constructional alloy steels. However, more than 0.0005to 0.004% boron produces no further increase in hardenability. Also, a trace of boron in-creases strength of low-carbon, plain molybdenum (0.40%) steel.

1.34 SECTION ONE

Chromium improves strength, hardenability, abrasion resistance, and resistance to at-mospheric corrosion. However, it reduces weldability. With small amounts of chromium,low-alloy steels have higher creep strength than carbon steels and are used where higherstrength is needed for elevated-temperature service. Also chromium is an important constit-uent of stainless steels.

Columbium in very small amounts produces relatively larger increases in yield point butsmaller increases in tensile strength of carbon steel. However, the notch toughness of thicksections is appreciably reduced.

Copper in amounts up to about 0.35% is very effective in improving the resistance ofcarbon steels to atmospheric corrosion. Improvement continues with increases in coppercontent up to about 1% but not so rapidly. Copper increases strength, with a proportionateincrease in fatigue limit. Copper also increases hardenability, with only a slight decrease inductility and little effect on notch toughness and weldability. However, steels with more than0.60% copper are susceptible to precipitation hardening. And steels with more than about0.5% copper often experience hot shortness during hot working, and surface cracks or rough-ness develop. Addition of nickel in an amount equal to about half the copper content iseffective in maintaining surface quality.

Hydrogen, which may be absorbed during steelmaking, embrittles steels. Ductility willimprove with aging at room temperature as the hydrogen diffuses out of the steel, fasterfrom thin sections than from thick. When hydrogen content exceeds 0.0005%, flaking, in-ternal cracks or bursts, may occur when the steel cools after rolling, especially in thicksections. In carbon steels, flaking may be prevented by slow cooling after rolling, to permitthe hydrogen to diffuse out of the steel.

Manganese increases strength, hardenability, fatigue limit, notch toughness, and corrosionresistance. It lowers the ductility and fracture transition temperatures. It hinders aging. Also,it counteracts hot shortness due to sulfur. For this last purpose, the manganese content shouldbe three to eight times the sulfur content, depending on the type of steel. However, man-ganese reduces weldability.

Molybdenum increases yield strength, hardenability, abrasion resistance, and corrosionresistance. It also improves weldability. However, it has an adverse effect on toughness andtransition temperature. With small amounts of molybdenum, low-alloy steels have highercreep strength than carbon steels and are used where higher strength is needed for elevated-temperature service.

Nickel increases strength, hardenability, notch toughness, and corrosion resistance. It isan important constituent of stainless steels. It lowers the ductility and fracture transitiontemperatures, and it reduces weldability.

Nitrogen increases strength, but it may cause aging. It also raises the ductility and fracturetransition temperatures.

Oxygen, like nitrogen, may be a cause of aging. Also, oxygen decreases ductility andnotch toughness.

Phosphorus increases strength, fatigue limit, and hardenability, but it decreases ductilityand weldability and raises the ductility transition temperature. Additions of aluminum, how-ever, improve the notch toughness of phosphorus-bearing steels. Phosphorus improves thecorrosion resistance of steel and works very effectively together with small amounts ofcopper toward this result.

Silicon increases strength, notch toughness, and hardenability. It lowers the ductility tran-sition temperature, but it also reduces weldability. Silicon often is used as a deoxidizer insteelmaking (see Art. 1.24).

Sulfur, which enters during the steelmaking process, can cause hot shortness. This resultsfrom iron sulfide inclusions, which soften and may rupture when heated. Also, the inclusionsmay lead to brittle failure by providing stress raisers from which fractures can initiate. Andhigh sulfur contents may cause porosity and hot cracking in welding unless special precau-tions are taken. Addition of manganese, however, can counteract hot shortness. It formsmanganese sulfide, which is more refractory than iron sulfide. Nevertheless, it usually isdesirable to keep sulfur content below 0.05%.


Titanium increases creep and rupture strength and abrasion resistance. It plays an im-portant role in preventing aging. It sometimes is used as a deoxidizer in steelmaking (seeArt. 1.24) and grain-growth inhibitor (see Art. 1.21).

Tungsten increases creep and rupture strength, hardenability and abrasion resistance. Itis used in steels for elevated-temperature service.

Vanadium, in amounts up to about 0.12%, increases rupture and creep strength withoutimpairing weldability or notch toughness. It also increases hardenability and abrasion resis-tance. Vanadium sometimes is used as a deoxidizer in steelmaking (see Art. 1.24) and grain-growth inhibitor (see Art. 1.21).

In practice, carbon content is limited so as not to impair ductility, notch toughness, andweldability. To obtain high strength, therefore, resort is had to other strengthening agentsthat improve these desirable properties or at least do not impair them as much as carbon.Often, the better these properties are required to be at high strengths, the more costly thesteels are likely to be.

Attempts have been made to relate chemical composition to weldability by expressingthe relative influence of chemical content in terms of carbon equivalent. One widely usedformula, which is a supplementary requirement in ASTM A6 for structural steels, is

Mn (Cr � Mo � V) (Ni � Cu)C � C � � � (1.17)eq 6 5 15

where C � carbon content, %Mn � manganese content, %Cr � chromium content, %

Mo � molybdenum, %V � vanadium, %

Ni � nickel content, %Cu � copper, %

Carbon equivalent is related to the maximum rate at which a weld and adjacent platemay be cooled after welding, without underbead cracking occurring. The higher the carbonequivalent, the lower will be the allowable cooling rate. Also, use of low-hydrogen weldingelectrodes and preheating becomes more important with increasing carbon equivalent. (Struc-tural Welding Code—Steel, American Welding Society, Miami, Fla.)

Though carbon provides high strength in steels economically, it is not a necessary ingre-dient. Very high strength steels are available that contain so little carbon that they are con-sidered carbon-free.

Maraging steels, carbon-free iron-nickel martensites, develop yield strengths from 150to 300 ksi, depending on alloying composition. As pointed out in Art. 1.20, iron-carbonmartensite is hard and brittle after quenching and becomes softer and more ductile whentempered. In contrast, maraging steels are relatively soft and ductile initially but becomehard, strong, and tough when aged. They are fabricated while ductile and later strengthenedby an aging treatment. These steels have high resistance to corrosion, including stress-corrosion cracking.


1.24 STEELMAKING METHODS

Structural steel is usually produced today by one of two production processes. In the tradi-tional process, iron or ‘‘hot metal’’ is produced in a blast furnace and then further processedin a basic oxygen furnace to make the steel for the desired products. Alternatively, steel canbe made in an electric arc furnace that is charged mainly with steel scrap instead of hot

1.36 SECTION ONE

metal. In either case, the steel must be produced so that undesirable elements are reducedto levels allowed by pertinent specifications to minimize adverse effects on properties.

In a blast furnace, iron ore, coke, and flux (limestone and dolomite) are charged intothe top of a large refractory-lined furnace. Heated air is blown in at the bottom and passedup through the bed of raw materials. A supplemental fuel such as gas, oil, or powdered coalis also usually charged. The iron is reduced to metallic iron and melted; then it is drawn offperiodically through tap holes into transfer ladles. At this point, the molten iron includesseveral other elements (manganese, sulfur, phosphorus, and silicon) in amounts greater thanpermitted for steel, and thus further processing is required.

In a basic oxygen furnace, the charge consists of hot metal from the blast furnace andsteel scrap. Oxygen, introduced by a jet blown into the molten metal, reacts with the im-purities present to facilitate the removal or reduction in level of unwanted elements, whichare trapped in the slag or in the gases produced. Also, various fluxes are added to reducethe sulfur and phosphorus contents to desired levels. In this batch process, large heats ofsteel may be produced in less than an hour.

An electric-arc furnace does not require a hot metal charge but relies mainly on steelscrap. The metal is heated by an electric arc between large carbon electrodes that projectthrough the furnace roof into the charge. Oxygen is injected to speed the process. This is aversatile batch process that can be adapted to producing small heats where various steelgrades are required, but it also can be used to produce large heats.

Ladle treatment is an integral part of most steelmaking processes. The ladle receivesthe product of the steelmaking furnace so that it can be moved and poured into either ingotmolds or a continuous casting machine. While in the ladle, the chemical composition of thesteel is checked, and alloying elements are added as required. Also, deoxidizers are addedto remove dissolved oxygen. Processing can be done at this stage to reduce further sulfurcontent, remove undesirable nonmetallics, and change the shape of remaining inclusions.Thus significant improvements can be made in the toughness, transverse properties, andthrough-thickness ductility of the finished product. Vacuum degassing, argon bubbling, in-duction stirring, and the injection of rare earth metals are some of the many procedures thatmay be employed.

Killed steels usually are deoxidized by additions to both furnace and ladle. Generally,silicon compounds are added to the furnace to lower the oxygen content of the liquid metaland stop oxidation of carbon (block the heat). This also permits addition of alloying elementsthat are susceptible to oxidation. Silicon or other deoxidizers, such as aluminum, vanadium,and titanium, may be added to the ladle to complete deoxidation. Aluminum, vanadium, andtitanium have the additional beneficial effect of inhibiting grain growth when the steel isnormalized. (In the hot-rolled conditions, such steels have about the same ferrite grain sizeas semikilled steels.) Killed steels deoxidized with aluminum and silicon (made to fine-grain practice) often are used for structural applications because of better notch toughnessand lower transition temperatures than semikilled steels of the same composition.


1.25 CASTING AND HOT ROLLING

Today, the continuous casting process is used to produce semifinished products directlyfrom liquid steel, thus eliminating the ingot molds and primary mills used previously. Withcontinuous casting, the steel is poured from sequenced ladles to maintain a desired level ina tundish above an oscillating water-cooled copper mold (Fig. 1.15). The outer skin of thesteel strand solidifies as it passes through the mold, and this action is further aided by watersprayed on the skin just after the strand exits the mold. The strand passes through sets ofsupporting rolls, curving rolls, and straightening rolls and is then rolled into slabs. The slabs


FIGURE 1.15 Schematic of slab caster.

are cut to length from the moving strand and held for subsequent rolling into finished product.Not only is the continuous casting process a more efficient method, but it also results inimproved quality through more consistent chemical composition and better surfaces on thefinished product.

Plates, produced from slabs or directly from ingots, are distinguished from sheet, strip,and flat bars by size limitations in ASTM A6. Generally, plates are heavier, per linear foot,than these other products. Plates are formed with straight horizontal rolls and later trimmed(sheared or gas cut) on all edges.

Slabs usually are reheated in a furnace and descaled with high-pressure water spraysbefore they are rolled into plates. The plastic slabs are gradually brought to desired dimen-sions by passage through a series of rollers. In the last rolling step, the plates pass throughleveling, or flattening, rollers. Generally, the thinner the plate, the more flattening required.After passing through the leveler, plates are cooled uniformly, then sheared or gas cut todesired length, while still hot.

Some of the plates may be heat treated, depending on grade of steel and intended use.For carbon steel, the treatment may be annealing, normalizing, or stress relieving. Plates ofHSLA or constructional alloy steels may be quenched and tempered. Some mills providefacilities for on-line heat treating or for thermomechanical processing (controlled rolling).Other mills heat treat off-line.

Shapes are rolled from continuously cast beam blanks or from blooms that first arereheated to 2250�F. Rolls gradually reduce the plastic blooms to the desired shapes and sizes.The shapes then are cut to length for convenient handling, with a hot saw. After that, theyare cooled uniformly. Next, they are straightened, in a roller straightener or in a gag press.Finally, they are cut to desired length, usually by hot shearing, hot sawing, or cold sawing.Also, column ends may be milled to close tolerances.

ASTM A6 requires that material for delivery ‘‘shall be free from injurious defects andshall have a workmanlike finish.’’ The specification permits manufacturers to condition plates

1.38 SECTION ONE

and shapes ‘‘for the removal of injurious surface imperfections or surface depressions bygrinding, or chipping and grinding. . . .’’ Except in alloy steels, small surface imperfectionsmay be corrected by chipping or grinding, then depositing weld metal with low-hydrogenelectrodes. Conditioning also may be done on slabs before they are made into other products.In addition to chipping and grinding, they may be scarfed to remove surface defects.

Hand chipping is done with a cold chisel in a pneumatic hammer. Machine chipping maybe done with a planer or a milling machine.

Scarfing, by hand or machine, removes defects with an oxygen torch. This can createproblems that do not arise with other conditioning methods. When the heat source is removedfrom the conditioned area, a quenching effect is produced by rapid extraction of heat fromthe hot area by the surrounding relatively cold areas. The rapid cooling hardens the steel,the amount depending on carbon content and hardenability of the steel. In low-carbon steels,the effect may be insignificant. In high-carbon and alloy steels, however, the effect may besevere. If preventive measures are not taken, the hardened area will crack. To prevent scarfingcracks, the steel should be preheated before scarfing to between 300 and 500�F and, in somecases, postheated for stress relief. The hardened surface later can be removed by normalizingor annealing.

Internal structure and many properties of plates and shapes are determined largely by thechemistry of the steel, rolling practice, cooling conditions after rolling, and heat treatment,where used. Because the sections are rolled in a temperature range at which steel is austenitic(see Art. 1.20), internal structure is affected in several ways.

The final austenitic grain size is determined by the temperature of the steel during thelast passes through the rolls (see Art. 1.21). In addition, inclusions are reoriented in thedirection of rolling. As a result, ductility and bendability are much better in the longitudinaldirection than in the transverse, and these properties are poorest in the thickness direction.

The cooling rate after rolling determines the distribution of ferrite and the grain size ofthe ferrite. Since air cooling is the usual practice, the final internal structure and, therefore,the properties of plates and shapes depend principally on the chemistry of the steel, sectionsize, and heat treatment. By normalizing the steel and by use of steels made to fine-grainpractice (with grain-growth inhibitors, such as aluminum, vanadium, and titanium), grainsize can be refined and properties consequently improved.

In addition to the preceding effects, rolling also may induce residual stresses in platesand shapes (see Art. 1.15). Still other effects are a consequence of the final thickness of thehot-rolled material.

Thicker material requires less rolling, the finish rolling temperature is higher, and thecooling rate is slower than for thin material. As a consequence, thin material has a superiormicrostructure. Furthermore, thicker material can have a more unfavorable state of stressbecause of stress raisers, such as tiny cracks and inclusions, and residual stresses.

Consequently, thin material develops higher tensile and yield strengths than thick materialof the same steel chemistry. ASTM specifications for structural steels recognize this usuallyby setting lower yield points for thicker material. A36 steel, however, has the same yieldpoint for all thicknesses. To achieve this, the chemistry is varied for plates and shapes andfor thin and thick plates. Thicker plates contain more carbon and manganese to raise theyield point. This cannot be done for high-strength steels because of the adverse effect onnotch toughness, ductility, and weldability.

Thin material generally has greater ductility and lower transition temperatures than thickmaterial of the same steel. Since normalizing refines the grain structure, thick material im-proves relatively more with normalizing than does thin material. The improvement is evengreater with silicon-aluminum-killed steels.



1.26 EFFECTS OF PUNCHING HOLES AND SHEARING

Excessive cold working of exposed edges of structural-steel members can cause embrittle-ment and cracking and should be avoided. Punching holes and shearing during fabricationare cold-working operations that can cause brittle failure in thick material.

Bolt holes, for example, may be formed by drilling, punching, or punching followed byreaming. Drilling is preferable to punching, because punching drastically coldworks the ma-terial at the edge of a hole. This makes the steel less ductile and raises the transition tem-perature. The degree of embrittlement depends on type of steel and plate thickness. Fur-thermore, there is a possibility that punching can produce short cracks extending radiallyfrom the hole. Consequently, brittle failure can be initiated at the hole when the member isstressed.

Should the material around the hole become heated, an additional risk of failure is intro-duced. Heat, for example, may be supplied by an adjacent welding operation. If the tem-perature should rise to the 400 to 850�F range, strain aging will occur in material susceptibleto it. The result will be a loss in ductility.

Reaming a hole after punching can eliminate the short, radial cracks and the risks ofembrittlement. For that purpose, the hole diameter should be increased from 1⁄16 to 1⁄4 in byreaming, depending on material thickness and hole diameter.

Shearing has about the same effects as punching. If sheared edges are to be left exposed.1⁄16 in or more material, depending on thickness, should be trimmed, usually by grinding ormachining. Note also that rough machining, for example, with edge planers making a deepcut, can produce the same effects as shearing or punching.

(M. E. Shank, Control of Steel Construction to Avoid Brittle Failure, Welding ResearchCouncil, New York.)

1.27 EFFECTS OF WELDING

Failures in service rarely, if ever, occur in properly made welds of adequate design.If a fracture occurs, it is initiated at a notchlike defect. Notches occur for various reasons.

The toe of a weld may form a natural notch. The weld may contain flaws that act as notches.A welding-arc strike in the base metal may have an embrittling effect, especially if weldmetal is not deposited. A crack started at such notches will propagate along a path determinedby local stresses and notch toughness of adjacent material.

Preheating before welding minimizes the risk of brittle failure. Its primary effect initiallyis to reduce the temperature gradient between the weld and adjoining base metal. Thus, thereis less likelihood of cracking during cooling and there is an opportunity for entrapped hy-drogen, a possible source of embrittlement, to escape. A consequent effect of preheating isimproved ductility and notch toughness of base and weld metals, and lower transition tem-perature of weld.

Rapid cooling of a weld can have an adverse effect. One reason that arc strikes that donot deposit weld metal are dangerous is that the heated metal cools very fast. This causessevere embrittlement. Such arc strikes should be completely removed. The material shouldbe preheated, to prevent local hardening, and weld metal should be deposited to fill thedepression.

Welding processes that deposit weld metal low in hydrogen and have suitable moisturecontrol often can eliminate the need for preheat. Such processes include use of low-hydrogenelectrodes and inert-arc and submerged-arc welding.

Pronounced segregation in base metal may cause welds to crack under certain fabricatingconditions. These include use of high-heat-input electrodes and deposition of large beads at

1.40 SECTION ONE

slow speeds, as in automatic welding. Cracking due to segregation, however, is rare for thedegree of segregation normally occurring in hot-rolled carbon-steel plates.

Welds sometimes are peened to prevent cracking or distortion, although special weldingsequences and procedures may be more effective. Specifications often prohibit peening ofthe first and last weld passes. Peening of the first pass may crack or punch through the weld.Peening of the last pass makes inspection for cracks difficult. Peening considerably reducestoughness and impact properties of the weld metal. The adverse effects, however, are elim-inated by the covering weld layer (last pass).

(M. E. Shank, Control of Steel Construction to Avoid Brittle Failure, Welding ResearchCouncil, New York; R. D. Stout and W. D. Doty, Weldability of Steels, Welding ResearchCouncil, New York.)

1.28 EFFECTS OF THERMAL CUTTING

Fabrication of steel structures usually requires cutting of components by thermal cuttingprocesses such as oxyfuel, air carbon arc, and plasma arc. Thermal cutting processes liberatea large quantity of heat in the kerf, which heats the newly generated cut surfaces to veryhigh temperatures. As the cutting torch moves away, the surrounding metal cools the cutsurfaces rapidly and causes the formation of a heat-affected zone analogous to that of a weld.The depth of the heat-affected zone depends on the carbon and alloy content of the steel,the thickness of the piece, the preheat temperature, the cutting speed, and the postheat treat-ment. In addition to the microstructural changes that occur in the heat-affected zone, the cutsurface may exhibit a slightly higher carbon content than material below the surface.

The detrimental properties of the thin layer can be improved significantly by using properpreheat, or postheat, or decreasing cutting speed, or any combination thereof. The hardnessof the thermally cut surface is the most important variable influencing the quality of thesurface as measured by a bend test. Plate chemistry (carbon content), Charpy V-notch tough-ness, cutting speed, and plate temperature are also important. Preheating the steel prior tocutting, and decreasing the cutting speed, reduce the temperature gradients induced by thecutting operation, thereby serving to (1) decrease the migration of carbon to the cut surface,(2) decrease the hardness of the cut surface, (3) reduce distortion, (4) reduce or give morefavorable distribution to the thermally induced stresses, and (5) prevent the formation ofquench or cooling cracks. The need for preheating increases with increased carbon and alloycontent of the steel, with increased thickness of the steel, and for cuts having geometriesthat act as high stress raisers. Most recommendations for minimum preheat temperatures aresimilar to those for welding.

The roughness of thermally cut surfaces is governed by many factors such as (1) unifor-mity of the preheat, (2) uniformity of the cutting velocity (speed and direction), and (3)quality of the steel. The larger the nonuniformity of these factors, the larger is the roughnessof the cut surface. The roughness of a surface is important because notches and stress raiserscan lead to fracture. The acceptable roughness for thermally cut surfaces is governed by thejob requirements and by the magnitude and fluctuation of the stresses for the particularcomponent and the geometrical detail within the component. In general, the surface rough-ness requirements for bridge components are more stringent than for buildings. The desiredmagnitude and uniformity of surface roughness can be achieved best by using automatedthermal cutting equipment where cutting speed and direction are easily controlled. Manualprocedures tend to produce a greater surface roughness that may be unacceptable for primarytension components. This is attributed to the difficulty in controlling both the cutting speedand the small transverse perturbations from the cutting direction.

(R. L. Brockenbrough and J. M. Barsom, Metallurgy, Chapter 1.1 in Constructional SteelDesign—An International Guide, R. Bjorhovde et al, Eds., Elsevier Science Publishers, Ltd.,New York.)

section 1 properties of structural steels …nguyen.hong.hai.free.fr/ebooks/science and...

Documents